LARGE-SCALE APPLICATIONS OF THE ACFC SYSTEM & COST ANALYSIS

Separation and recovery of volatile organic chemicals and solvents from the Synthetic Organic Chemical Manufacturing Industries (SOCMI), industrial painting processes, and water and wastewater treatment plants are potential areas of targeting application of the ACFC sorption system. This system can be considered for any industrial process exhausting sizable quantities of valuable toxic vapors that can be efficiently recovered by carbon adsorption. Utilization of this system with electrothermal regeneration makes the ACFC system a good substitute for conventional carbon adsorption systems when the waste hot gas or steam are not economically or technically a feasible option for the regeneration. In general, electrothermal regeneration shortens the desorption time and increases the regenerated concentration appreciably; thereby contributing to cost reduction of the condensation unit. Electrothermal regeneration is preferrable compared to steam because steam may contribute to polymerization reactions on the surface of adsorbant. This type of reaction is usually caused by the breakdown of some reactive organic compounds on the surface of carbon in the presence of transitional metals as catalysts (McInnes, 1995). In general, carbon adsorption is not recommended for VOC streams containing ketones. Continued exposure to ketones can produce exothermic reactions that can cause fires in conventional carbon adsorption systems. This reaction can take place in conventional carbon adsorption systems because transitional metals are available in commercial activated carbon adsorbents. Since the ACFC does not contain transitional metals (Table 3.8), the ACFC sorption system is expected to result in safe recovery of ketones. Feasibility of adsorption, desorption and condensation of acetone and MEK via ACFC were extensively demonstrated in the previous chapters. In the following section, design conditions and cost analysis of an ACFC sorption system are provided for recovery of acetone.

Preliminary Design and Economic Analysis of the ACFC Electrothermal Sorption System

The Net Present Value (NPV) and break-even price analysis of the ACFC electrothermal sorption system was done by Dr. Subhash Bhagwat at the Illinois State Geological Survey. The analysis is based on the flow diagram in Figure 3.1. It includes the adsorption-desorption and condensation stages. A separate analysis of the condensation stage is presented in Section 5..

Table 5.1 summarizes design conditions of the ACFC system. These values were used as input parameters in the economic analysis. The Producer Price Index (PPI) for Capital Equipment published by the U.S. Department of Commerce was used to update the investment data to 1996 status. The scenarios I and II differ from one another primarily due to the price of ACFC mass quoted by two different sources. The base case for the estimation of operating costs assumes that in an industrial setting the plant operations will be automated enough to permit this unit to function - operate and maintain - with 1 hour of labor per shift, 3 shifts per day, 365 days per year. This assumption is subject to a greater degree of uncertainty than the assumptions regarding the prices of electricity and LN₂ consumed in the process. The energy consumption for desorption was calculated to be 144 MJ/hr. The price of electricity for industrial use currently averages about 6 cents/kwh. Liquid nitrogen price of 1991 was updated to 1996 by using the Consumer Price Index (CPI). The operating costs are assumed to increase at a 3% annual rate but revenues from the sale of recovered acetone are assumed to be constant over the ten year period. This is realistic because market competition and alternative sources of chemicals often do not permit price increases.

Table 5.2 presents the NPV analysis for scenario II as the base case. Scenario II is based on the higher of the two price quotes for the ACFC mass and the adsorption cycle time of 2.6 hours. The mass of ACFC required in this case was calculated to be 112.5 kg per reactor vessel or 225 kg total. A reduction in adsorption time to 1 hour would reduce the reactor price from $1,750 to $1,000 per vessel and the mass of ACFC needed from 112.5 kg to 88 kg per vessel.

The NPV analysis determines the price of recovered acetone at which the process pays for itself. We have assumed that all the capital needed has been borrowed at 20 % annual interest rate. The annual cash flows were discounted also at the same rate of 20 %. The rate of 20 % is essential to attract funds to a new technology such as this. The risks assumed by the potential investor must be justified by a higher return on investment than the investment alternatives available to the investor.

The break-even economic analysis indicates that the recovered acetone must be saleable for at least $1.08 per kg. The current market value of acetone is $15.95 per gallon or about $5.40 per kg, assuming a specific gravity of 0.78. The process of adsorption, desorption and condensation thus promises to be highly economical.

Sensitivity Analysis

The economic outcome is highly sensitive to changes in components of the operating costs. For example, if the labor requirements increase from 1 person-hour per shift to 2 person-hours, the break-even price of recovered acetone increases by 26 cents to $1.34/kg. Similarly, an increase in electricity price from 6 cents to 10 cents per kwh raises the break-even price by 13 cents to $1.21/kg. The effect is cumulative if both costs go up at the same time.

The process economics are also highly sensitive to the price of the ACFC mass. The break-even price of acetone declines from $1.08/kg to $0.91/kg if the price of the ACFC mass declines from $120 to $20 per kg. A reduction in adsorption cycle time from 2.6 hour to 1 hour makes the process less sensitive to the price of ACFC mass. At 1 hour adsorption cycle time and an ACFC price of $20/kg, the break-even price of recovered acetone is $0.89/kg. It rises to $0.93/kg if the price of ACFC mass rises to $120/kg. This price is 15 cents/kg lower than the base case with an adsorption cycle time of 2.6 hours. Finally, the lowering of the interest and discount rates from 20 to 15% results in a 10% decline in break-even acetone price from $1.08 to $0.97 per kg.

In conclusion, the NPV analysis of the entire system including adsorption, desorption and condensation indicates excellent prospects for an economically profitable process of recovering acetone. The sensitivity analysis shows the effect of component cost changes on the total break-even price of recovered acetone. The components of the annual operating cost are more important than the initial investment level in influencing the total cost. Efforts to reduce the labor requirements for operation and maintenance and the energy required for desorption would pay highly. The cost of ACFC, although treated here as investment, is a variable of considerable significance. Reducing the adsorption cycle time will not only reduce total cost but also make the total cost less sensitive to changes in ACFC price. Overall, the preliminary economic analysis offers several opportunities for future process development and optimization. A detailed economic analysis with refined data and a comprehensive sensitivity analysis is recommended. Caution should be exercised in transposing the results to other TVOCs because each substance will have a different response to the process.

Preliminary Design and Economic Analysis of the Condensation System

Large-scale condenser design can be carried out by first determining the process gas stream characteristics such as TVOC vapor concentration, temperature and gas flow rate. By assuming equilibrium conditions, the desired condenser temperature can be determined from the vapor concentration dependence on temperature at the desired outlet concentration (Wagner equation eq. 3.2). Once the temperature is known, an appropriate refrigerant can be selected (e.g. Table 3.5). Then by modeling the axial concentration profile, the appropriate surface area and condenser length can be determined from the condensing surface required to reach the desired outlet vapor concentration (mass transfer model, Appendix D).

Shell-and-tube condensers range in size from under 1 m² to as large as 30,000 m² in surface area (Kern, 1950). Kern (1950) provides standard design criteria for the construction of shell-and tube condensers. A common type of condenser is the fixed tube sheet, in which a bundle of inside tubes are encased in an outside shell (Figure 5.1). This type of shell-and-tube condenser will be used for the following analysis, in conjunction with the general design methodology outlined above.

Figure 5.1 Large-scale schematic of a tubesheet shell-and-tube ndirect-contact condenser

 

The process conditions are the same as given in Table 5.1. Desorption is assumed to decrease the flow rate by an order of magnitude and increase the acetone concentration to the saturation vapor concentration at 294 K (as was shown with the bench-scale system in Section 4). Therefore the condenser inlet stream has the following characteristics:

Inlet vapor concentration to adsorber = 26% by volume acetone (from Wagner equation)

Carrier gas = N₂ gas (approximates air)

Inlet flow rate to adsorber = 1 m³/min

Temperature = 294 K

Pressure = 1 atm

To achieve a 99% mass removal efficiency, the outlet acetone concentration must equal 0.24% by volume. The saturation vapor concentration vs. temperature graph shows that the condenser must be cooled to 220 K (Figure 3.3). The refrigerant selected to achieve this condenser temperature is LN₂ (Table 3.5).

Gas stream characteristics are input in the mass transfer/thermodynamic model to determine condensation surface area, mass of LN₂ required and mass of acetone condensed. The results of the model show that a conservative surface area of 20 m² is needed to achieve the desired outlet acetone concentration (Figure 5.2). A safety factor of 1.2 is applied to assure that sufficient surface area is available and to account for heat transfer reduction due to condenser fouling (e.g., water vapor). The design surface area required is 24 m². Mass flow rate of LN₂ is predicted to be 980 kg/day to condense 890 kg/day of acetone (Table 5.4). Kern (1950) can be used to optimize the geometric configuration to achieve the required 24 m² of condensation surface area depending on site specific considerations.

An economic analysis can be conducted based on required condensation surface area. The estimated condenser cost for an 8 ft tube length, 14 BWG fixed tube sheet condenser with 24 m² (262 ft²) of heat transfer surface is approximately $8,000 (Figure 5.3). Other capital costs are also estimated using the USEPA derived cost factors (USEPA^b, 1991). Cost analysis shows that an estimated $73,100 (1996 dollars) capital investment is required for equipment and installation (Table 5.5). Amortized over 10 years at a 10% interest rate, the cost is $11,900/yr (1996 dollars).

Figure 5.3 Costs for fixed tubesheet condensers. BWG is Birmingham wire gage; 14 BWG is a 0.216 cm tube wall thickness (USEPA, 1991).

The overall recovered acetone credits for the system analyzed in the chapter is $494,400/yr (1996 dollars) resulting in a recovery of $3.92/kg of acetone condensed. Capital costs only attributed approximately $11,900/yr to the overall annual costs. The largest cost factor was expenditures for LN₂ at $26,600/yr. Recovery credits result in an additional income of $561,341/yr.